Diese Dissertation beschreibt die Entwicklung und Anwendung des Klimawirkungsmoduls des ICLIPS-Modells, eines integrierten Modells des Klimawandels ('Integrated Assessment'-Modell). Vorangestellt ist eine Diskussion des gesellschaftspolitischen Kontexts, in dem modellbasiertes 'Integrated Assessment' stattfindet, aus der wichtige Anforderungen an die Spezifikation des Klimawirkungsmoduls abgeleitet werden. <br /> <br /> Das 'Integrated Assessment' des Klimawandels umfasst eine weiten Bereich von Aktivitäten zur wissenschaftsbasierten Unterstützung klimapolitischer Entscheidungen. Hierbei wird eine Vielzahl von Ansätzen verfolgt, um politikrelevante Informationen über die erwarteten Auswirkungen des Klimawandels zu berücksichtigen. Wichtige Herausforderungen in diesem Bereich sind die große Bandbreite der relevanten räumlichen und zeitlichen Skalen, die multifaktorielle Verursachung vieler 'Klimafolgen', erhebliche wissenschaftliche Unsicherheiten sowie die Mehrdeutigkeit unvermeidlicher Werturteile. Die Entwicklung eines hierarchischen Konzeptmodells erlaubt die Strukturierung der verschiedenen Ansätze sowie die Darstellung eines mehrstufigen Entwicklungsprozesses, der sich in der Praxis und der zu Grunde liegenden Theorie von Studien zur Vulnerabilität hinsichtlich des Klimawandels wiederspiegelt. <br /> <br /> 'Integrated Assessment'-Modelle des Klimawandels sind wissenschaftliche Werkzeuge, welche eine vereinfachte Beschreibung des gekoppelten Mensch-Klima-Systems enthalten. Die wichtigsten entscheidungstheoretischen Ansätze im Bereich des modellbasierten 'Integrated Assessment' werden im Hinblick auf ihre Fähigkeit zur adäquaten Darstellung klimapolitischer Entscheidungsprobleme bewertet. Dabei stellt der 'Leitplankenansatz' eine 'inverse' Herangehensweise zur Unterstützung klimapolitischer Entscheidungen dar, bei der versucht wird, die Gesamtheit der klimapolitischen Strategien zu bestimmen, die mit einer Reihe von zuvor normativ bestimmten Mindestkriterien (den sogenannten 'Leitplanken') verträglich sind. Dieser Ansatz verbindet bis zu einem gewissen Grad die wissenschaftliche Strenge und Objektivität simulationsbasierter Ansätze mit der Fähigkeit von Optimierungsansätzen, die Gesamtheit aller Entscheidungsoptionen zu berücksichtigen. Das ICLIPS-Modell ist das erste 'Integrated Assessment'-Modell des Klimawandels, welches den Leitplankenansatz implementiert. <br /> <br /> Die Darstellung von Klimafolgen ist eine wichtige Herausforderung für 'Integrated Assessment'-Modelle des Klimawandels. Eine Betrachtung bestehender 'Integrated Assessment'-Modelle offenbart große Unterschiede in der Berücksichtigung verschiedener vom Klimawandel betroffenen Sektoren, in der Wahl des bzw. der Indikatoren zur Darstellung von Klimafolgen, in der Berücksichtigung nicht-klimatischer Entwicklungen einschließlich gezielter Anpassungsmaßnahmen an den Klimawandel, in der Behandlung von Unsicherheiten und in der Berücksichtigung von 'singulären' Ereignissen. 'Integrated Assessment'-Modelle, die auf einem Inversansatz beruhen, stellen besondere Anforderungen an die Darstellung von Klimafolgen. Einerseits muss der Detaillierungsgrad hinreichend sein, um Leitplanken für Klimafolgen sinnvoll definieren zu können; andererseits muss die Darstellung effizient genug sein, um die Gesamtheit der möglichen klimapolitischen Strategien erkunden zu können. Großräumige Singularitäten können häufig durch vereinfachte dynamische Modelle abgebildet werden. Diese Methode ist jedoch weniger geeignet für reguläre Klimafolgen, bei denen die Bestimmung relevanter Ergebnisse in der Regel die Berücksichtigung der Heterogenität von klimatischen, naturräumlichen und sozialen Faktoren auf der lokalen oder regionalen Ebene erfordert. <br /> <br /> Klimawirkungsfunktionen stellen sich als die geeignetste Darstellung regulärer Klimafolgen im ICLIPS-Modell heraus. Eine Klimawirkungsfunktion beschreibt in aggregierter Form die Reaktion eines klimasensitiven Systems, wie sie von einem geographisch expliziten Klimawirkungsmodell für eine repräsentative Teilmenge möglicher zukünftiger Entwicklungen simuliert wurde. Die in dieser Arbeit vorgestellten Klimawirkungsfunktionen nutzen die globale Mitteltemperatur sowie die atmosphärische CO2-Konzentration als Prädiktoren für global und regional aggregierte Auswirkungen des Klimawandels auf natürliche Ökosysteme, die landwirtschaftliche Produktion und die Wasserverfügbarkeit. Die Anwendung einer 'Musterskalierungstechnik' ermöglicht hierbei die Berücksichtigung der regionalen und saisonalen Muster des Klimaänderungssignals aus allgemeinen Zirkulationsmodellen, ohne die Effizienz der dynamischen Modellkomponenten zu beeinträchtigen. <br /> <br /> Bemühungen zur quantitativen Abschätzung zukünftiger Klimafolgen sehen sich bei der Wahl geeigneter Indikatoren in der Regel einem Zielkonflikt zwischen der Relevanz eines Indikators für Entscheidungsträger und der Zuverlässigkeit, mit der dieser bestimmt werden kann, gegenüber. Eine Reihe von nichtmonetären Indikatoren zur aggregierten Darstellung von Klimafolgen in Klimawirkungsfunktionen wird präsentiert, welche eine Balance zwischen diesen beiden Zielen anstreben und gleichzeitig die Beschränkungen berücksichtigen, die sich aus anderen Komponenten des ICLIPS-Modells ergeben. Klimawirkungsfunktionen werden durch verschiedene Typen von Diagrammen visualisiert, welche jeweils unterschiedliche Perspektiven auf die Ergebnismenge der Klimawirkungssimulationen erlauben.<br /> <br /> Die schiere Anzahl von Klimawirkungsfunktionen verhindert ihre umfassende Darstellung in dieser Arbeit. Ausgewählte Ergebnisse zu Veränderungen in der räumlichen Ausdehnung von Biomen, im landwirtschaftlichen Potential verschiedener Länder und in der Wasserverfügbarkeit in mehreren großen Einzugsgebieten werden diskutiert. Die Gesamtheit der Klimawirkungsfunktionen wird zugänglich gemacht durch das 'ICLIPS Impacts Tool', eine graphische Benutzeroberfläche, die einen bequemen Zugriff auf über 100.000 Klimawirkungsdiagramme ermöglicht. Die technischen Aspekte der Software sowie die zugehörige Datenbasis wird beschrieben. <br /> <br /> Die wichtigste Anwendung von Klimawirkungsfunktionen ist im 'Inversmodus', wo sie genutzt werden, um Leitplanken zur Begrenzung von Klimafolgen in gleichzeitige Randbedingungen für Variablen aus dem optimierenden ICLIPS-Klima-Weltwirtschafts-Modell zu übersetzen. Diese Übersetzung wird ermöglicht durch Algorithmen zur Bestimmung von Mengen erreichbarer Klimazustände ('reachable climate domains') sowie zur parametrisierten Approximation zulässiger Klimafenster ('admissible climate windows'), die aus Klimawirkungsfunktionen abgeleitet werden. Der umfassende Bestand an Klimawirkungsfunktionen zusammen mit diesen Algorithmen ermöglicht es dem integrierten ICLIPS-Modell, in flexibler Weise diejenigen klimapolitischen Strategien zu bestimmen, welche bestimmte in biophysikalischen Einheiten ausgedrückte Begrenzungen von Klimafolgen explizit berücksichtigen. Diese Möglichkeit bietet kein anderes intertemporal optimierendes 'Integrated Assessment'-Modell. Eine Leitplankenanalyse mit dem integrierten ICLIPS-Modell unter Anwendung ausgewählter Klimawirkungsfunktionen für Veränderungen natürlicher Ökosysteme wird beschrieben. In dieser Analyse werden so genannte 'notwendige Emissionskorridore' berechnet, die vorgegebene Beschränkungen hinsichtlich der maximal zulässigen globalen Vegetationsveränderungen und der regionalen Klimaschutzkosten berücksichtigen. Dies geschieht sowohl für eine 'Standardkombination' der drei gewählten Kriterien als auch für deren systematische Variation. <br /> <br /> Eine abschließende Diskussion aktueller Entwicklungen in der 'Integrated Assessment'-Modellierung stellt diese Arbeit mit anderen einschlägigen Bemühungen in Beziehung. / This thesis describes the development and application of the impacts module of the ICLIPS model, a global integrated assessment model of climate change. The presentation of the technical aspects of this model component is preceded by a discussion of the sociopolitical context for model-based integrated assessments, which defines important requirements for the specification of the model.<br /> <br /> Integrated assessment of climate change comprises a broad range of scientific efforts to support the decision-making about objectives and measures for climate policy, whereby many different approaches have been followed to provide policy-relevant information about climate impacts. Major challenges in this context are the large diversity of the relevant spatial and temporal scales, the multifactorial causation of many climate impacts', considerable scientific uncertainties, and the ambiguity associated with unavoidable normative evaluations. A hierarchical framework is presented for structuring climate impact assessments that reflects the evolution of their practice and of the underlying theory.<br /> <br /> Integrated assessment models of climate change (IAMs) are scientific tools that contain simplified representations of the relevant components of the coupled society-climate system. The major decision-analytical frameworks for IAMs are evaluated according to their ability to address important aspects of the pertinent social decision problem. The guardrail approach is presented as an inverse' framework for climate change decision support, which aims to identify the whole set of policy strategies that are compatible with a set of normatively specified constraints (guardrails'). This approach combines, to a certain degree, the scientific rigour and objectivity typical of predictive approaches with the ability to consider virtually all decision options that is at the core of optimization approaches. The ICLIPS model is described as the first IAM that implements the guardrail approach.<br /> <br /> The representation of climate impacts is a key concern in any IAM. A review of existing IAMs reveals large differences in the coverage of impact sectors, in the choice of the impact numeraire(s), in the consideration of non-climatic developments, including purposeful adaptation, in the handling of uncertainty, and in the inclusion of singular events. IAMs based on an inverse approach impose specific requirements to the representation of climate impacts. This representation needs to combine a level of detail and reliability that is sufficient for the specification of impact guardrails with the conciseness and efficiency that allows for an exploration of the complete domain of plausible climate protection strategies. Large-scale singular events can often be represented by dynamic reduced-form models. This approach, however, is less appropriate for regular impacts where the determination of policy-relevant results generally needs to consider the heterogeneity of climatic, environmental, and socioeconomic factors at the local or regional scale.<br /> <br /> Climate impact response functions (CIRFs) are identified as the most suitable reduced-form representation of regular climate impacts in the ICLIPS model. A CIRF depicts the aggregated response of a climate-sensitive system or sector as simulated by a spatially explicit sectoral impact model for a representative subset of plausible futures. In the CIRFs presented here, global mean temperature and atmospheric CO2 concentration are used as predictors for global and regional impacts on natural vegetation, agricultural crop production, and water availability. Application of a pattern scaling technique makes it possible to consider the regional and seasonal patterns in the climate anomalies simulated by several general circulation models while ensuring the efficiency of the dynamic model components.<br /> <br /> Efforts to provide quantitative estimates of future climate impacts generally face a trade-off between the relevance of an indicator for stakeholders and the exactness with which it can be determined. A number of non-monetary aggregated impact indicators for the CIRFs is presented, which aim to strike the balance between these two conflicting goals while taking into account additional constraints of the ICLIPS modelling framework. Various types of impact diagrams are used for the visualization of CIRFs, each of which provides a different perspective on the impact result space.<br /> <br /> The sheer number of CIRFs computed for the ICLIPS model precludes their comprehensive presentation in this thesis. Selected results referring to changes in the distribution of biomes in different biogeographical regions, in the agricultural potential of various countries, and in the water availability in selected major catchments are discussed. The full set of CIRFs is accessible via the ICLIPS Impacts Tool, a graphical user interface that provides convenient access to more than 100,000 impact diagrams developed for the ICLIPS model. The technical aspects of the software are described as well as the accompanying database of CIRFs.<br /> <br /> The most important application of CIRFs is in inverse' mode, where they are used to translate impact guardrails into simultaneous constraints for variables from the optimizing ICLIPS climate-economy model. This translation is facilitated by algorithms for the computation of reachable climate domains and for the parameterized approximation of admissible climate windows derived from CIRFs. The comprehensive set of CIRFs, together with these algorithms, enables the ICLIPS model to flexibly explore sets of climate policy strategies that explicitly comply with impact guardrails specified in biophysical units. This feature is not found in any other intertemporally optimizing IAM. A guardrail analysis with the integrated ICLIPS model is described that applies selected CIRFs for ecosystem changes. So-called necessary carbon emission corridors' are determined for a default choice of normative constraints that limit global vegetation impacts as well as regional mitigation costs, and for systematic variations of these constraints.<br /> <br /> A brief discussion of recent developments in integrated assessment modelling of climate change connects the work presented here with related efforts.

O problema das mudanças climáticas globais somente pode ser resolvido através de um longo processo de coordenação política internacional no qual os principais atores são os governos dos Estados Nacionais. O Protocolo de Quioto é o primeiro acordo internacional para controlar as emissões de gases de efeito estufa, e muitas outras rodadas de negociações ocorrerão ao longo deste século gerando novos acordos com o mesmo objetivo. Inúmeros fatores, relacionados de forma complexa, têm influenciado e influenciarão os resultados desses futuros acordos. Dentre os principais estão as incertezas sobre os vários aspectos, físicos, biológicos, econômicos e políticos, do problema das mudanças climáticas. O objetivo deste trabalho foi construir um modelo de análise integrada, que tornasse transparente os principais elementos e elos da cascata de incertezas existente no problema das mudanças climáticas e a sua influência nos resultados daquelas negociações e nos permitisse fazer simulações exploratórias sobre os efeitos da evolução dessas incertezas sobre os resultados da seqüência de negociações após o Protocolo de Quioto. O modelo representa de forma estilizada importantes elementos que participam da estrutura de decisão coletiva sobre abatimento de reduções de emissões e em alguns aspectos é mais realista que vários modelos existentes sobre tomada de decisões sobre mudança de clima. Ele é composto de um módulo que representa o sistema do clima, um módulo que representa as economias nacionais, um módulo que representa os tomadores de decisões governamentais e um módulo que representa as negociações sobre reduções de emissões. Os tomadores de decisões são representados como agentes que têm planos de desenvolvimento econômico para seus países e modelos sobre como o clima global e regional e as suas economias funcionam, o que lhes permite fazerem projeções futuras do aquecimento global, das suas economias e emissões, e do impacto que a mudança de clima produzirá em seu território. Dessa forma podem estimar o impacto que o aquecimento global poderá ter sobre seus planos de desenvolvimento. Essas análises são feitas dentro de um horizonte de antecipação que depende do grau de incertezas na época em que as projeções são realizadas. Representamos algumas das principais incertezas na ciência do clima e análise econômica do problema através de distribuições de probabilidades de certos parâmetros chave, tais como a sensibilidade climática e a difusividade térmica do Oceano, que podem variar ao longo do tempo. Existe uma infinidade de cenários possíveis de evolução dessas incertezas, mas somente alguns com significados intuitivos importantes. Uma negociação sobre cotas de reduções de emissões é representada como um jogo não-cooperativo, e o acordo entre os países é um equilíbrio de Nash desse jogo. Cada governo, antes de ir à mesa de negociações, baseado nestas projeções e nas possíveis ações dos outros governos, elabora suas estratégias sobre o quanto abater de suas emissões. O resultado final de um jogo é influenciado pelas distribuições de probabilidades que representam as incertezas da época em que cada agente faz suas projeções. Algumas distribuições em algumas épocas podem revelar aos agentes a probabilidade de impactos climáticos perigosos em suas economias, influenciando fortemente as suas escolhas de abatimento e o resultados das negociações. Esse ciclo de projeções-análise-negociação se repete várias vezes ao longo do tempo definindo uma seqüência de acordos de reduções de emissões, um conseqüente aquecimento global e uma distribuição de impactos regionais ao longo do mesmo período. A comparação das trajetórias finais de aquecimento e impactos gerados para cada cenário escolhido de evolução das incertezas ajuda-nos a compreender as possíveis evoluções das futuras negociações sobre abatimento de emissões. Implementamos uma versão simplificada do modelo em um programa de computador através da plataforma Microsoft Excel, considerando somente dois blocos de países. Com esse protótipo, é possível simular cenários de evolução das incertezas, representando evoluções possíveis da ciência do clima e a análise econômica do problema, e com isso estudar a influência dessas evoluções nos resultados de uma seqüência de negociações sobre cotas de reduções de emissões entre dois blocos de países, que uma vez definidos os valores de certos parâmetros, podem representar os países do Anexo I e Não-Anexo I da Convenção Quadro de Mudança de Clima das Nações Unidas. Nesse caso o Protocolo de Quioto é o primeiro acordo de uma série de acordos ao longo do Século XXI. O modelo foi construído apoiando-se na literatura tradicional de modelagem integrada em mudanças climáticas, mas inova a pesquisa nesta área em vários pontos importantes. Dentre eles está a variação gradual de várias incertezas e a incorporação de uma interpretação do conceito de impactos climáticos perigosos conforme o Artigo 2 da Convenção Quadro e seu papel nos resultados das negociações. A linguagem e a plataforma utilizada nos permite comunicar de forma clara os elos da cascata de incertezas e sua influência sobre as negociações, tanto a cientistas, quanto a analistas econômicos e políticos envolvidos com o tema. A plataforma utilizada torna também o programa aberto para análises críticas e modificações a uma larga classe de analistas do tema, além de poder ser utilizado no ensino, que ainda carece de ferramentas pedagógicas que permitam uma ampla divulgação desse tema a não especialistas. Apesar de ser um modelo sobre mudanças climáticas, os vários módulos têm uma estrutura conceitual que lhes permite serem adaptados para tratar outras questões ambientais globais e regionais onde as soluções baseiam-se em tomada de decisão coletiva negociada e sob incertezas. / The problem of the global climate change only can to be solved through a long term process of international political coordination which the principal actors are the governments of the National States. The Kyoto Protocol is the first international agreement to control the greenhouse gas (GHG) emissions, and the others rounds of negotiations will happen through this century producing new deals with the same goal. A number of factors related of a complex way, have influenced and will influence the results of these future agreements. Among the principal factors, are the uncertainties about the physical, biological, ecological, economical and poltical aspects of the global climate change problem. The goal of this work was to contruct a integrated assessment model, that show in a transparent way some the majors elements and links of the uncertainties cascade that exist in the problem of the climate change, and their influence on the results on that negotiations. The model permit us to make exploratory simulations about the effects of the evolution that uncertainties in the results of the sequence of negotiations after the Kyoto Protocol. The model represent in a stylized way important elements that participate in the structure of collective decision making about abatement of GHG emissions and in some aspects is more realistic that several existent models on decision making about climate change. It is composed of one module that represent the climate system, one module that represent the national economies, one module that represent the governmental decision makers and one module that represent the negotiations about reductions of GHG emissions. The climate model include a carbon cycle and an energetic balance simple models with white noise in the radiative forcing to repesent the natural climate variability. The economic models include abrupt changes in the damages functions. The The decision makers are represented as agents that have economical development plans for their countries and have models of the climate change and their national economies. These models permit to each one make projections on future global warming, economical growth, emissions and the climate impacts on each territory. Thus each agent can estimate the impact of the global warming on his development plan under a action of abatement of emissions. These analysis are made with the aid of a antecipation horizon that depends of the uncertainty levels in the epoch which the projections are made. We represented some of majors uncertainties in the climate science and economic analysis on the problem through probabilities distributions of certain key parameters, such that the climate sensibility and the ocean thermal diffusivity, which can to vary in time. There is a number of possible evolution scenarios for these uncertainties, but only a few with important intuitive means. We take only these more important scenarios to simulation. A negotiation on reduction quotas of emissions is represented as a non-cooperative game and the agreement between countries is a Nash equilibrium of the game. Each government before to negotiate choose his strategies on how much abate emissions based on his projections. The game final result is influenced by the probabilities distributions that represent the uncertainties in the epoch of the negotiations. Some distributions can reveal high probabilities for dangerous climate impacts in their economies, influencing strongly the choices of emission abatements and the results of negotiations. The cycle of projection-analysisnegotiation can to repeat many times defining a sequence of agreements and then a emission trajectory, and as consequence a global warming and distributions of regional impacts. The comparison of final trajectories of warming and impacts for each negotiations sequence help us to understand the possible evolutions of future negotiations and his role in the climate change problem.

In nie dagewesener Größenordnung greift der Mensch durch die Verbrennung fossiler Energieträger und der weiträumigen Umgestaltung der Landoberfläche in die globale Umwelt ein. Klimawandel und Übernutzung natürlicher Ressourcen könnten schon in diesem Jahrhundert die Anpassungsfähigkeiten vieler ökologischer und sozialer Systeme übersteigen und somit zu Konflikten und politischer Destabilisierung führen. Vor diesem Hintergrund soll diese Studie zu einem besseren Verständnis der wichtigsten globalen Triebkräfte beitragen, die die Entwicklung der terrestrischen Biosphäre in diesem Jahrhundert prägen werden: Klimawandel und menschliche Landnutzung. Auf der Basis eines Dynamischen Globalen Vegetationsmodells werden im ersten Teil der vorliegenden Arbeit zwei große klimatische Störungen des globalen Kohlenstoffkreislaufs untersucht, die innerhalb der letzten drei Jahrzehnte beobachtet wurden. Im Fordergrund steht die Frage, wie sich die Veränderungen von Temperatur-, Niederschlags- und Strahlungsbedingungen auf pflanzliche Produktivität und Zersetzungsprozesse im Boden auswirkten. Es zeigt sich, dass vermehrte Kohlenstoffspeicherung in der Landbiosphäre den überwiegenden Teil der atmosphärischen CO2 Anomalien erklärt. Der zweite Teil dieser Arbeit beschäftigt sich mit der weltweit steigenden Nachfrage nach Bioenergie, die aufgrund des flächenintensiven Anbaus von Biomasse zur wichtigsten Triebkraft für zukünftige Landnutzungsänderungen werden könnte. Aus der Kombination von Vegetationsmodellierung und räumlichen Datenanalysen werden globale Bioenergiepotentiale unter Berücksichtigung verschiedener Nachhaltigkeitsanforderungen bestimmt und mögliche ökologische Auswirkungen des großräumigen Anbaus von Energiepflanzen abgeschätzt. Im Jahr 2050 könnten demnach 15-25% des weltweiten Energiebedarfs durch Bioenergie abgedeckt werden. Dafür müssten allerdings natürliche Ökosysteme in großem Umfang in Agrarland umgewandelt werden. / Human activities, primarily the combustion of fossil fuels and the global modification of the land surface, are transforming the Earth System at unprecedented scale. Climate change and the overexploitation of natural resources may soon overwhelm the adaptive capacities of many ecosystems and societies, which could lead to substantial losses in human well-being and political destabilization. In this context, it is the goal of this thesis to contribute to a better understanding of the most important global drivers that will determine the future of the land biosphere during this century: climate change and human land use. Based on a Dynamic Global Vegetation Model (DGVM), the first part of this thesis examines two large climatic disturbances of the terrestrial carbon cycle that were observed during the last three decades. These analyses focus on the effects of changes in temperature, precipitation and radiation on plant productivity and soil decomposition. Results indicate that increased carbon storage in the land biosphere explains the most part of the atmospheric CO2 anomaly. The second part of this thesis addresses the worldwide increasing demand for bioenergy that may become the most important driver of future land use change due to the large area requirements of biomass cultivation. A combination of vegetation modeling and spatial data analyses is used to assess global bioenergy potentials that consider various sustainability requirements for food security, biodiversity protection and the reduction of greenhouse gas emissions and to evaluate the environmental impacts of large-scale energy crop cultivation. The results indicate that bioenergy may provide between 15 and 25% of the global energy demand in 2050. Exploiting these potentials, however, requires the conversion of large amounts of natural vegetation into agricultural land affecting a large number of ecosystems already fragmented and degraded by land use change.

Do changes in how cross-equatorial energy transport is partitioned between the ocean and atmosphere impact the hemispheric climate response to forcing? To find out, we alter the cross-equatorial ocean heat transport in a state-of-the-art GCM and ascertain how changes in energy transport and its partitioning impact hemispheric climate and precipitation sensitivity following abrupt CO2-doubling. We further evaluate the applicability our results in CMIP6-class ESMs, where AMOC facilitates the northward cross-equatorial ocean heat transport. In our experiments, changes in ocean cross-equatorial energy transport trigger compensating changes in atmospheric energy transport through changes in the Hadley cells and a shift in the Intertropical Convergence Zone. However, the climate sensitivity in each hemisphere is linearly related to the ocean heat transport convergence, not atmospheric energy transport convergence, due to the impact of ocean heating on evaporation and atmospheric specific humidity. Similarly, we also find that ocean heat transport convergence controls the hemispheric precipitation sensitivity through the impact of ocean heating on surface evaporation. This relationship is also evident in CMIP6 models, where we find differences in hemispheric precipitation sensitivity to be related to the Atlantic Meridional Overturning Circulation (AMOC). Changes in the AMOC control hemispheric differences in upper ocean heat content, which then affect how the hydrologic cycle responds to CO2 forcing in each hemisphere. These results suggest that ocean dynamics impact the hemispheric climate response to CO2 forcing, particularly how much regional precipitation changes with warming. / Graduate